Abstract: In this paper we address the issue of classifying the fluorescent intensity of a sample in Indirect Immuno-Fluorescence (IIF). Since IIF is a subjective, semi-quantitative test in its very nature, we discuss a strategy to reliably label the image data set by using the diagnoses performed by different physicians. Then, we discuss image pre-processing, feature extraction and selection. Finally, we propose two ANN-based classifiers that can separate intrinsically dubious samples and whose error tolerance can be flexibly set. Measured performance shows error rates less than 1%, which candidates the method to be used in daily medical practice either to perform pre-selection of cases to be examined, or to act as a second reader.
Abstract: Extended Kalman Filter (EKF) is probably the most
widely used estimation algorithm for nonlinear systems. However,
not only it has difficulties arising from linearization but also many
times it becomes numerically unstable because of computer round off
errors that occur in the process of its implementation. To overcome
linearization limitations, the unscented transformation (UT) was
developed as a method to propagate mean and covariance
information through nonlinear transformations. Kalman filter that
uses UT for calculation of the first two statistical moments is called
Unscented Kalman Filter (UKF). Square-root form of UKF (SRUKF)
developed by Rudolph van der Merwe and Eric Wan to
achieve numerical stability and guarantee positive semi-definiteness
of the Kalman filter covariances. This paper develops another
implementation of SR-UKF for sequential update measurement
equation, and also derives a new UD covariance factorization filter
for the implementation of UKF. This filter is equivalent to UKF but
is computationally more efficient.
Abstract: In this paper, a neural tree (NT) classifier having a
simple perceptron at each node is considered. A new concept for
making a balanced tree is applied in the learning algorithm of the
tree. At each node, if the perceptron classification is not accurate and
unbalanced, then it is replaced by a new perceptron. This separates
the training set in such a way that almost the equal number of patterns
fall into each of the classes. Moreover, each perceptron is trained only
for the classes which are present at respective node and ignore other
classes. Splitting nodes are employed into the neural tree architecture
to divide the training set when the current perceptron node repeats
the same classification of the parent node. A new error function based
on the depth of the tree is introduced to reduce the computational
time for the training of a perceptron. Experiments are performed to
check the efficiency and encouraging results are obtained in terms of
accuracy and computational costs.
Abstract: Wireless location is to determine the mobile station (MS) location in a wireless cellular communications system. When fewer base stations (BSs) may be available for location purposes or the measurements with large errors in non-line-of-sight (NLOS) environments, it is necessary to integrate all available heterogeneous measurements to achieve high location accuracy. This paper illustrates a hybrid proposed schemes that combine time of arrival (TOA) at three BSs and angle of arrival (AOA) information at the serving BS to give a location estimate of the MS. The proposed schemes mitigate the NLOS effect simply by the weighted sum of the intersections between three TOA circles and the AOA line without requiring a priori information about the NLOS error. Simulation results show that the proposed methods can achieve better accuracy when compare with Taylor series algorithm (TSA) and the hybrid lines of position algorithm (HLOP).
Abstract: This study is concerned with the investigation of the
suitability of several empirical and semi-empirical drying models
available in the literature to define drying behavior of viscose yarn
bobbins. For this purpose, firstly, experimental drying behaviour of
viscose bobbins was determined on an experimental dryer setup
which was designed and manufactured based on hot-air bobbin
dryers used in textile industry. Afterwards, drying models considered
were fitted to the experimentally obtained moisture ratios. Drying
parameters were drying temperature and bobbin diameter. The fit
was performed by selecting the values for constants in the models in
such a way that these values make the sum of the squared differences
between the experimental and the model results for moisture ratio
minimum. Suitability of fitting was specified as comparing the
correlation coefficient, standard error and mean square deviation.
The results show that the most appropriate model in describing the
drying curves of viscose bobbins is the Page model.
Abstract: As in today's semiconductor industries test costs can make up to 50 percent of the total production costs, an efficient test error detection becomes more and more important. In this paper, we present a new machine learning approach to test error detection that should provide a faster recognition of test system faults as well as an improved test error recall. The key idea is to learn a classifier ensemble, detecting typical test error patterns in wafer test results immediately after finishing these tests. Since test error detection has not yet been discussed in the machine learning community, we define central problem-relevant terms and provide an analysis of important domain properties. Finally, we present comparative studies reflecting the failure detection performance of three individual classifiers and three ensemble methods based upon them. As base classifiers we chose a decision tree learner, a support vector machine and a Bayesian network, while the compared ensemble methods were simple and weighted majority vote as well as stacking. For the evaluation, we used cross validation and a specially designed practical simulation. By implementing our approach in a semiconductor test department for the observation of two products, we proofed its practical applicability.
Abstract: Majority of researches conducted on Iranian urban
development plans indicate that they have been almost unsuccessful
in terms of draft, execution and goal achievement. Lack or shortage
of essential statistics and information can be listed as an important
reason of the failure of these plans. Lack of figures and information
has turned into an obvious part of the country-s statistics officials.
This problem has made urban planner themselves to embark on
physical surveys including real estate and land pricing, population
and economic census of the city. Apart from the problems facing
urban developers, the possibility of errors is high in such surveys.
In the present article, applying the interview technique, it has
been mentioned that utilizing multipurpose cadastre system as a land
information system is essential for urban development plans in Iran.
It can minimize or even remove the failures facing urban
development plans.
Abstract: In this paper we use exponential particle swarm
optimization (EPSO) to cluster data. Then we compare between
(EPSO) clustering algorithm which depends on exponential variation
for the inertia weight and particle swarm optimization (PSO)
clustering algorithm which depends on linear inertia weight. This
comparison is evaluated on five data sets. The experimental results
show that EPSO clustering algorithm increases the possibility to find
the optimal positions as it decrease the number of failure. Also show
that (EPSO) clustering algorithm has a smaller quantization error
than (PSO) clustering algorithm, i.e. (EPSO) clustering algorithm
more accurate than (PSO) clustering algorithm.
Abstract: Testing is an activity that is required both in the
development and maintenance of the software development life cycle
in which Integration Testing is an important activity. Integration
testing is based on the specification and functionality of the software
and thus could be called black-box testing technique. The purpose of
integration testing is testing integration between software
components. In function or system testing, the concern is with overall
behavior and whether the software meets its functional specifications
or performance characteristics or how well the software and
hardware work together. This explains the importance and necessity
of IT for which the emphasis is on interactions between modules and
their interfaces. Software errors should be discovered early during
IT to reduce the costs of correction. This paper introduces a new type
of integration error, presenting an overview of Integration Testing
techniques with comparison of each technique and also identifying
which technique detects what type of error.
Abstract: It is observed that the Weighted least-square (WLS)
technique, including the modifications, results in equiripple error
curve. The resultant error as a percent of the ideal value is highly
non-uniformly distributed over the range of frequencies for which the
differentiator is designed. The present paper proposes a modification
to the technique so that the optimization procedure results in lower
maximum relative error compared to the ideal values. Simulation
results for first order as well as higher order differentiators are given
to illustrate the excellent performance of the proposed method.
Abstract: A concern that researchers usually face in different
applications of Artificial Neural Network (ANN) is determination of
the size of effective domain in time series. In this paper, trial and
error method was used on groundwater depth time series to determine
the size of effective domain in the series in an observation well in
Union County, New Jersey, U.S. different domains of 20, 40, 60, 80,
100, and 120 preceding day were examined and the 80 days was
considered as effective length of the domain. Data sets in different
domains were fed to a Feed Forward Back Propagation ANN with
one hidden layer and the groundwater depths were forecasted. Root
Mean Square Error (RMSE) and the correlation factor (R2) of
estimated and observed groundwater depths for all domains were
determined. In general, groundwater depth forecast improved, as
evidenced by lower RMSEs and higher R2s, when the domain length
increased from 20 to 120. However, 80 days was selected as the
effective domain because the improvement was less than 1% beyond
that. Forecasted ground water depths utilizing measured daily data
(set #1) and data averaged over the effective domain (set #2) were
compared. It was postulated that more accurate nature of measured
daily data was the reason for a better forecast with lower RMSE
(0.1027 m compared to 0.255 m) in set #1. However, the size of input
data in this set was 80 times the size of input data in set #2; a factor
that may increase the computational effort unpredictably. It was
concluded that 80 daily data may be successfully utilized to lower the
size of input data sets considerably, while maintaining the effective
information in the data set.
Abstract: In this paper we present a noise reduction filter for video processing. It is based on the recently proposed two dimensional steering kernel, extended to three dimensions and further augmented to suit the spatial-temporal domain of video processing. Two alternative filters are proposed - the time symmetric kernel and the time asymmetric kernel. The first reduces the noise on single sequences, but to handle the problems at scene shift the asymmetric kernel is introduced. The performance of both are tested on simulated data and on a real video sequence together with the existing steering kernel. The proposed kernels improves the Rooted Mean Squared Error (RMSE) compared to the original steering kernel method on video material.
Abstract: As the majority of faults are found in a few of its
modules so there is a need to investigate the modules that are
affected severely as compared to other modules and proper
maintenance need to be done in time especially for the critical
applications. As, Neural networks, which have been already applied
in software engineering applications to build reliability growth
models predict the gross change or reusability metrics. Neural
networks are non-linear sophisticated modeling techniques that are
able to model complex functions. Neural network techniques are
used when exact nature of input and outputs is not known. A key
feature is that they learn the relationship between input and output
through training. In this present work, various Neural Network Based
techniques are explored and comparative analysis is performed for
the prediction of level of need of maintenance by predicting level
severity of faults present in NASA-s public domain defect dataset.
The comparison of different algorithms is made on the basis of Mean
Absolute Error, Root Mean Square Error and Accuracy Values. It is
concluded that Generalized Regression Networks is the best
algorithm for classification of the software components into different
level of severity of impact of the faults. The algorithm can be used to
develop model that can be used for identifying modules that are
heavily affected by the faults.
Abstract: In this paper, a PSO-based approach is proposed to
derive a digital controller for redesigned digital systems having an interval plant based on resemblance of the extremal gain/phase
margins. By combining the interval plant and a controller as an interval system, extremal GM/PM associated with the loop transfer function
can be obtained. The design problem is then formulated as an optimization problem of an aggregated error function revealing the deviation on the extremal GM/PM between the redesigned digital
system and its continuous counterpart, and subsequently optimized by
a proposed PSO to obtain an optimal set of parameters for the digital controller. Computer simulations have shown that frequency
responses of the redesigned digital system having an interval plant bare a better resemblance to its continuous-time counter part by the incorporation of a PSO-derived digital controller in comparison to those obtained using existing open-loop discretization methods.
Abstract: Due to the high percentage of induction motors in industrial market, there exist a large opportunity for energy savings. Replacement of working induction motors with more efficient ones can be an important resource for energy savings. A calculation of energy savings and payback periods, as a result of such a replacement, based on nameplate motor efficiency or manufacture-s data can lead to large errors [1]. Efficiency of induction motors (IMs) can be extracted using some procedures that use the no-load test results. In the cases that we must estimate the efficiency on-line, some of these procedures can-t be efficient. In some cases the efficiency estimates using the rating values of the motor, but these procedures can have errors due to the different working condition of the motor. In this paper the efficiency of an IM estimated by using the genetic algorithm. The results are compared with the measured values of the torque and power. The results show smaller errors for this procedure compared with the conventional classical procedures, hence the cost of the equipments is reduced and on-line estimation of the efficiency can be made.
Abstract: In this paper, the application of neural networks to study the design of short-term temperature forecasting (STTF) Systems for Kermanshah city, west of Iran was explored. One important architecture of neural networks named Multi-Layer Perceptron (MLP) to model STTF systems is used. Our study based on MLP was trained and tested using ten years (1996-2006) meteorological data. The results show that MLP network has the minimum forecasting error and can be considered as a good method to model the STTF systems.
Abstract: The recent global financial problem urges government
to play role in stimulating the economy due to the fact that private
sector has little ability to purchase during the recession. A concerned
question is whether the increased government spending crowds out
private consumption and whether it helps stimulate the economy. If
the government spending policy is effective; the private consumption
is expected to increase and can compensate the recent extra
government expense. In this study, the government spending is
categorized into government consumption spending and government
capital spending. The study firstly examines consumer consumption
along the line with the demand function in microeconomic theory.
Three categories of private consumption are used in the study. Those
are food consumption, non food consumption, and services
consumption. The dynamic Almost Ideal Demand System of the three
categories of the private consumption is estimated using the Vector
Error Correction Mechanism model. The estimated model indicates
the substituting effects (negative impacts) of the government
consumption spending on budget shares of private non food
consumption and of the government capital spending on budget share
of private food consumption, respectively. Nevertheless the result
does not necessarily indicate whether the negative effects of changes
in the budget shares of the non food and the food consumption means
fallen total private consumption. Microeconomic consumer demand
analysis clearly indicates changes in component structure of
aggregate expenditure in the economy as a result of the government
spending policy. The macroeconomic concept of aggregate demand
comprising consumption, investment, government spending (the
government consumption spending and the government capital
spending), export, and import are used to estimate for their
relationship using the Vector Error Correction Mechanism model.
The macroeconomic study found no effect of the government capital
spending on either the private consumption or the growth of GDP
while the government consumption spending has negative effect on
the growth of GDP. Therefore no crowding out effect of the
government spending is found on the private consumption but it is
ineffective and even inefficient expenditure as found reducing growth
of the GDP in the context of Thailand.
Abstract: This paper presents the applicability of artificial
neural networks for 24 hour ahead solar power generation forecasting
of a 20 kW photovoltaic system, the developed forecasting is suitable
for a reliable Microgrid energy management. In total four neural
networks were proposed, namely: multi-layred perceptron, radial
basis function, recurrent and a neural network ensemble consisting in
ensemble of bagged networks. Forecasting reliability of the proposed
neural networks was carried out in terms forecasting error
performance basing on statistical and graphical methods. The
experimental results showed that all the proposed networks achieved
an acceptable forecasting accuracy. In term of comparison the neural
network ensemble gives the highest precision forecasting comparing
to the conventional networks. In fact, each network of the ensemble
over-fits to some extent and leads to a diversity which enhances the
noise tolerance and the forecasting generalization performance
comparing to the conventional networks.
Abstract: Electrocardiogram (ECG) segmentation is necessary
to help reduce the time consuming task of manually annotating
ECG-s. Several algorithms have been developed to segment the ECG
automatically. We first review several of such methods, and then
present a new single lead segmentation method based on Adaptive
piecewise constant approximation (APCA) and Piecewise derivative
dynamic time warping (PDDTW). The results are tested on the QT
database. We compared our results to Laguna-s two lead method. Our
proposed approach has a comparable mean error, but yields a slightly
higher standard deviation than Laguna-s method.
Abstract: In this paper we introduce a novel kernel classifier
based on a iterative shrinkage algorithm developed for compressive
sensing. We have adopted Bregman iteration with soft and hard
shrinkage functions and generalized hinge loss for solving l1 norm
minimization problem for classification. Our experimental results
with face recognition and digit classification using SVM as the
benchmark have shown that our method has a close error rate
compared to SVM but do not perform better than SVM. We have
found that the soft shrinkage method give more accuracy and in some
situations more sparseness than hard shrinkage methods.